ecommerce-amazon-market-product-search

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and displays Amazon marketplace data (such as product titles and descriptions), creating a surface for potential indirect prompt injection attacks where malicious instructions could be embedded in listing content.\n
  • Ingestion points: External data enters via the API response handled by scripts/amazon_market_product_search.py.\n
  • Boundary markers: The skill documentation recommends using jq to selectively read saved data, which serves as a boundary to prevent the entire untrusted payload from entering the agent's primary reasoning context.\n
  • Capability inventory: The skill includes a Python script capable of making network requests via urllib and writing results to the filesystem.\n
  • Sanitization: The skill lacks explicit sanitization or filtering of the retrieved marketplace data strings before they are presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:13 AM
Security Audit — agent-trust-hub — ecommerce-amazon-market-product-search